Fujun Liu

Water in a river flows around rocks, air bends around an airplane wing: these motions are described by equations that are hard to solve even on a powerful computer. Neural networks can approximate such solutions, but they often err when data is collected at only a few points or is noisy.

A pot of boiling water bubbles and churns, mixing everything inside. Take it off the heat, and the motion subsides as the water slowly cools. The Universe is like that: right after the Big Bang, it was a searing 'soup' of particles that constantly collided and transformed into one another. But as space expanded, the cosmos cooled, and particles met each other less and less often.